The proposed case-control study is designed to assess the relation of long- term oral contraceptive use to the incidence of choriocarcinoma and invasive hydatidiform mole, the most serous manifestations of gestational trophoblastic disease. A total of 340 cases (90 choriocarcinomas and 250 invasive moles), admitted to any of six regional centers for the treatment of trophoblastic disease or registered by the Massachusetts Tumor Registry, will be studied. One control, identified through random-digit dialing, will be enrolled for each case. Controls will be matched to the cases on geography, age at the time of the most recent pregnancy (or earliest molar pregnancy) and calendar year of that pregnancy. Information on the factors of interest and factors that may confound or modify their relation to risk will be obtained by personal interview. Detailed data on the timing and duration of oral contraceptive use will be collected. In addition, data will be obtained on several measures of sexual activity so that sexual activity, a marker for exposure to infectious agents, can be evaluated as an independent risk factor and controlled as a potential confounder. Invasive mole and choriocarcinoma are serious diseases affecting young women and requiring treatment by chemotherapy. Although the incidence is low in North America and Europe, it is higher in other parts of the world such as Southeast Asia. Almost nothing is known about the etiology of these diseases. Research efforts in the past have focused on treatment; epidemiologic studies have been few and most have been small. In the course of making multiple comparisons in a case-control study of several cancers, we observed an unexpected association between long-term oral contraceptive use and gestational trophoblastic disease. Since then, an association with oral contraceptive use was observed in two recently completed case-control studies, one a study of prevalent cases of choriocarcinoma in the United States, and the other, a study of incident cases of invasive mole and complete hydatidiform mole in China. The latter findings underscore the importance of conducting the proposed study, which will permit a thorough exploration of the relation between oral contraceptive use and trophoblastic disease, as well as providing information on other suspected risk factors such as body size, parity, ABO blood group, and carotene intake.